A

Altera

FPGA and SoC provider with software

Principal Agentic AI Architect

Natural Language Processing & LLMsConfirmed live in the last 24 hours
Full-timeShanghai, ChinaNo salary listedIn-person

About the Job

Requirements

  • A Master's or Bachelor's degree and 10+ years of experience in AI/ML systems, distributed systems, or hardware–software co-design.
  • Recognized expertise in AI and Large Language Models (LLMs).
  • Deep understanding of LLM architectures and inference, Retrieval-Augmented Generation (RAG) systems and vector databases, and agentic AI or autonomous agents.
  • Hands-on experience building production-grade LLM applications, multi-step reasoning and orchestration workflows, and scalable and reliable AI systems in production.
  • Strong background in FPGA, ASIC, or EDA systems and heterogeneous computing architectures.
  • Familiarity with hardware acceleration for AI workloads and performance optimization across the system stack.
  • Experience designing distributed systems and APIs and microservices architecture, and deploying cloud and/or edge AI.
  • Hands-on experience with LLM fine-tuning, quantization, and model distillation; on-device or edge AI deployment; and robotics, autonomous systems, or real-time AI pipelines.
  • Knowledge of compiler stacks such as MLIR and TVM, and AI model optimization for FPGA platforms.
  • Experience with Python; C++ is preferred.

Responsibilities

  • Design and implement agentic AI systems using LLMs, tools, and multi-agent orchestration.
  • Architect RAG pipelines for enterprise-scale knowledge integration.
  • Define end-to-end AI system architecture spanning data ingestion, reasoning, and action.
  • Develop workflows using tool-using agents, planning and reasoning agents, and multi-agent collaboration frameworks.
  • Evaluate and integrate leading open-source and proprietary LLMs.
  • Optimize prompt engineering, memory, and reasoning capabilities.
  • Co-design AI systems with FPGA acceleration, ASIC platforms, and heterogeneous CPU, GPU, and FPGA compute.
  • Explore LLM inference optimization on FPGA, including quantization, compilation, and partitioning.
  • Collaborate with hardware teams on EDA and system integration.
  • Build internal platforms using LangChain, LlamaIndex, and custom agent orchestration frameworks.
  • Develop reusable components for memory management, tool integration, and knowledge retrieval.
  • Lead architecture reviews, guide engineering teams, and mentor senior engineers.
  • Influence cross-functional teams and drive innovation in AI and hardware convergence.

Desired Qualifications

  • C++ programming experience.

About the Company

Altera

FPGA and SoC provider with software

Altera designs and manufactures FPGAs and SoCs, along with software tools for hardware design and deployment. Customers can configure these programmable chips after fabrication to fit specific applications, helping them reach market faster and optimize performance. The company differentiates itself by offering a broad configurable silicon platform and a dedicated design software suite as an independent provider after its separation from Intel. Its goal is to deliver high-performance, adaptable processing solutions for industries like automotive, aerospace, communications, AI, cloud, and edge computing.

Company size
Not listed
Company stage
IPO
Headquarters
San Jose, CA
Founded
1983

Recently Posted Jobs

Explore more opportunities at Altera

See more jobs